Sodium channel blockers and class III antiarrhythmic compounds, as well as beta blockers, have been used in preventing recurrences of sudden cardiac death. In recent years, implantable cardioverter-defibrillators (ICDs) have been used increasingly, but no data from randomized trials comparing antiarrhythmic drug and ICD therapy have been reported in this setting. In 1987, the Cardiac Arrest Study Hamburg (CASH), a prospective, randomized trial, was initiated to compare metoprolol, amiodarone, propafenone, and ICD implantation in patients surviving sudden cardiac death due to documented ventricular tachycardia and/or ventricular fibrillation. The details of the study design and preliminary results are presented herein. The primary endpoint of the study is total mortality. The data reviewed in March 1992, representing a mean follow-up period of 11 months, indicated no significant differences among patients randomized to metoprolol, amiodarone, and ICDs. However, there was a significantly higher total mortality and cardiac arrest recurrence in patients randomized to propafenone compared with those randomized to the ICD treatment limb. The study continues with the deletion of the propafenone treatment limb.